Sample Solvent and Eluent

[mdyo]

Hi everyone,
I have read that sample solvent should be miscible with the eluent as immiscibility of the two may cause separation problems. However, I am using a method in which the sample solvent (hexane) is immiscible with the eluent (MeCN/ Water) but I have not noticed any problem in the separation.

Has anyone faced any problem because of such immiscibility?
What kind of problems might be caused by this immiscibility?

Any comments are very appreciated

Many thanks in advance.

[Uwe Neue]

If the injection volume is small, it may not be a problem. The hexane slowly elute, as it becomes soluble. If you inject a very large volume, you will see the trouble immediately, since the pressure will shoot up. However, even then it is not a fundamental issue, sine it can be purged out of the column again.

[mdyo]

Thanks a lot Uwe

The injection volume is 10 ul and I am using a C18 250x4.6 mm column. I am trying to validate this method which has been modified from an original GC method but I am worry about the immiscibility of the two solvents.

Do you think this will not be an argumental issue for the validation?

Thank you again.

[tom jupille]

That's what your validation is all about: to demonstrate that your method is suitable for the intended purpose. If you can show that it works, it works.

[james little]

I used to inject samples in solvents that weren't miscible with the organic phase.

As long as you injected a small enough volume, it was OK.

After injecting, I would often see a large increase in the backpressure of system which would immediately drop back to normal in ~5 seconds. Assumed some was precipitating on frit?

I used a 0.5 micron precolumn filter/holder, and if I plugged the system, could replace the frit to recover..

[HW Mueller]

James, could the pressure spike be due to surface tension phenomena, ... dewetting maybe. Seems that I have seen that.

[Uwe Neue]

The pressure spike is normal, due to the injection of immiscible solvents. If the injection volume is small enough, it is not a big deal, if you make a large injection, you may sit there for a while to try to clean up the mess. However, the 5 seconds in James's case was not bad.

[ljc]

If the peak of interest elutes late enough, and you demonstrate that the solvent doesn't coelute with it, you should be OK. In fact, with early eluting components, even a miscible "strong" solvent can broaden or shoulder your peaks ( for example dissolving a polar analyte in straight MeOH or ACN will often give a shouldered or split peak).

If your method passes all the validation criteria, you are home free.